OSSForks/pytorch
1
0
Fork 0
mirror of https://github.com/zebrajr/pytorch.git synced 2025-12-07 12:21:27 +01:00
Code Issues Actions 38 Packages Projects Releases Wiki Activity

Enabling GPU device runs for operators (#23461)

Browse Source 
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/23461

Enabling GPU device runs for production operator shapes.

Reviewed By: xw285cornell, mingzhe09088

Differential Revision: D16526928

fbshipit-source-id: 46657963f4b0bc43d14205ccf1b63d588657e388
This commit is contained in:
Abhinav Jauhri 2019-07-26 18:50:22 -07:00 committed by Facebook Github Bot
parent 23e526e6ff
commit ffef0e03b7
1 changed files with 36 additions and 35 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

71
benchmarks/operator_benchmark/benchmark_test_generator.py
View File

@ -4,7 +4,7 @@ from __future__ import print_function
from __future__ import unicode_literals
from collections import namedtuple
import copy
import copy
from benchmark_core import TestConfig
from benchmark_caffe2 import register_caffe2_op_test_case
from benchmark_pytorch import register_pytorch_op_test_case
@ -13,12 +13,12 @@ from benchmark_utils import SkipInputShape
def _generate_test(configs, bench_op, OperatorTestCase, run_backward, op_name_function=None):
"""Generate PyTorch/Caffe2 tests of operators with different inputs.
Args:
Args:
configs: a dictionary that has the input shapes
bench_op: a subclass of Caffe2BenchmarkBase/TorchBenchmarkBase which includes tensor
creation and operator execution
OperatorTestCase: a named tuple to save the metadata of an test
run_backward: a bool parameter indicating backward path
OperatorTestCase: a named tuple to save the metadata of an test
run_backward: a bool parameter indicating backward path
op_name_function: a dictionary includes operator name and function
"""
for config in configs:
@ -26,10 +26,10 @@ def _generate_test(configs, bench_op, OperatorTestCase, run_backward, op_name_fu
tags = None
for attr in config:
# tags is only used in our benchmark backend to filter tests and
# it will be removed from config which is then passed to the init function
# an example of config and atrr is:
# it will be removed from config which is then passed to the init function
# an example of config and atrr is:
# config: [{'M': 16}, {'N': 16}, {'K': 64}, {'tags': 'short'}]
# attr: {'tags': 'short'}
# attr: {'tags': 'short'}
if "tags" in attr:
tags = attr["tags"]
continue
@ -42,7 +42,7 @@ def _generate_test(configs, bench_op, OperatorTestCase, run_backward, op_name_fu
# op_name_function is a dictionary which has op_name and op_function.
# an example of op_name_function is:
# {'op_name' : 'abs', 'op_function' : torch.abs}
# op_function is concatenated with the input dict then passed to the init function
# op_function is concatenated with the input dict then passed to the init function
# op_name is passed to the set_module_name function
init_dict = copy.deepcopy(test_attrs)
if op_name_function is not None:
@ -64,42 +64,43 @@ def _generate_test(configs, bench_op, OperatorTestCase, run_backward, op_name_fu
OpMeta = namedtuple("OpMeta", "op_type num_inputs input_dims input_types \
output_dims num_outputs args")
output_dims num_outputs args device")
def generate_c2_test_from_ops(ops_metadata, bench_op, tags):
"""
This function is used to generate Caffe2 tests based on the meatdata
of operators. The metadata includes seven fields which are 1) op_type:
the name of the operator. 2) num_inputs: the number of input blobs.
of operators. The metadata includes seven fields which are 1) op_type:
the name of the operator. 2) num_inputs: the number of input blobs.
3) input_dims: a dictionary which includes the shapes of the input blobs.
4) input_types: a list which includes the types of input blobs. 5)
4) input_types: a list which includes the types of input blobs. 5)
output_dims: a dictionary which includes the shapes of output blobs.
6) num_oupts: the number of output blobs. 7) args: a dictionary which
includes the args for th operator.
includes the args for th operator.
Here is an example to show the metadata for the WeighedSum operator
op_type : WeightedSum
num_inputs: 4
num_inputs: 4
input_dims: {'0': [256], '1': [1], '2': [256], '3': [1]}
input_types: ['float', 'float', 'float', 'float']
output_dims: {'0': [256]}
num_outputs: 4
output_dims: {'0': [256]}
num_outputs: 4
args: {}
TODO(mingzhe0908): introduce device and add it to the benchmark name
TODO(mingzhe0908): introduce device and add it to the benchmark name
"""
for op_metadata in ops_metadata:
tmp_attrs = OpMeta(op_metadata.op_type,
op_metadata.num_inputs,
tmp_attrs = OpMeta(op_metadata.op_type,
op_metadata.num_inputs,
op_metadata.input_dims,
op_metadata.input_types,
op_metadata.output_dims,
op_metadata.num_outputs,
op_metadata.args)
op_metadata.input_types,
op_metadata.output_dims,
op_metadata.num_outputs,
op_metadata.args,
op_metadata.device)
test_attrs = tmp_attrs._asdict()
op = bench_op()
op.init(**test_attrs)
test_name = op.test_name("short")
input_config = "Shapes: {}, Type: {}, Args: {}".format(
op_metadata.input_dims,
op_metadata.input_dims,
op_metadata.input_types,
str(op_metadata.args))
test_config = TestConfig(test_name, input_config, tags, run_backward=False)
@ -112,34 +113,34 @@ def generate_c2_test_from_ops(ops_metadata, bench_op, tags):
def generate_pt_test(configs, pt_bench_op):
""" This function creates PyTorch op test based on the given operator
"""
_generate_test(configs, pt_bench_op, register_pytorch_op_test_case,
_generate_test(configs, pt_bench_op, register_pytorch_op_test_case,
run_backward=False)
def generate_c2_test(configs, c2_bench_op):
""" This function creates Caffe2 op test based on the given operator
""" This function creates Caffe2 op test based on the given operator
"""
_generate_test(configs, c2_bench_op, register_caffe2_op_test_case,
_generate_test(configs, c2_bench_op, register_caffe2_op_test_case,
run_backward=False)
def generate_pt_gradient_test(configs, pt_bench_op):
""" This function creates PyTorch op test based on the given operator
"""
_generate_test(configs, pt_bench_op, register_pytorch_op_test_case,
_generate_test(configs, pt_bench_op, register_pytorch_op_test_case,
run_backward=True)
def generate_c2_gradient_test(configs, c2_bench_op):
""" This function creates Caffe2 op test based on the given operator
""" This function creates Caffe2 op test based on the given operator
"""
_generate_test(configs, c2_bench_op, register_caffe2_op_test_case,
_generate_test(configs, c2_bench_op, register_caffe2_op_test_case,
run_backward=True)
def generate_pt_tests_from_op_list(ops_list, configs, pt_bench_op):
""" This function creates pt op tests one by one from a list of dictionaries.
ops_list is a list of dictionary. Each dictionary includes
ops_list is a list of dictionary. Each dictionary includes
the name of the operator and the math operation. Here is an example of using this API:
unary_ops_configs = op_bench.config_list(
attrs=[...],
@ -152,13 +153,13 @@ def generate_pt_tests_from_op_list(ops_list, configs, pt_bench_op):
],
)
class UnaryOpBenchmark(op_bench.TorchBenchmarkBase):
def init(self, M, N, op_name, op_func):
...
def init(self, M, N, op_name, op_func):
...
def forward(self):
...
op_bench.generate_pt_tests_from_op_list(unary_ops_list, unary_ops_configs, UnaryOpBenchmark)
"""
for op in ops_list:
_generate_test(configs, pt_bench_op, register_pytorch_op_test_case,
run_backward=False,
_generate_test(configs, pt_bench_op, register_pytorch_op_test_case,
run_backward=False,
op_name_function=op)